The impact and mechanism of neighbourhood social capital on mental health: a cross-sectional survey based on the floating elderly population in China

Table 5 Statistical results of mediating effects

Dependent variable	Happiness		Mental illness
Independent variable	Bonding neighbourhood social capital		Bridging neighbourhood social capital
Mediating variable	Physical exercise	Psychological integration	Social adaptation
a	0.424^c	0.663^c	0.221^c
b	0.112^a	0.327^c	−0.257^a
c’	0.105	0.105	−0.591^a
c	0.484^a	0.484^a	−0.545^a
m	0.048	0.217	−0.057
Sobel	1.977^a	4.852^c	−1.934^a
Monte Carlo	1.947^a	4.828^c	−1.882^a
RIT	0.313	0.674	0.088
RID	0.456	2.069	0.097

Note. a is the path coefficient between the independent variable and the mediating variable, b is the path coefficient between the mediating variable and the dependent variable, c is the path coefficient between the independent variable and the dependent variable (without the mediating variable), and c' is the path coefficient between the independent variable and the dependent variable after the mediating variable is added. Sobel statistical value =\(\frac{a \times b}{{\sqrt {a^{2} s_{b}^{2} + b^{2} s_{a}^{2} } }}\) The Monte Carlo value is a statistical value using the Sobel calculation method after several estimates. RIT represents the proportion of mediating effects in the total effect, while RID represents the ratio of mediating effects to direct effects. ^a and ^c indicate significance at the 5% and 0.1% levels (two-tailed)

ISSN: 1471-2318